Ohio       Summary of Green Recount Team reports of county recount efforts-  only about half of counties have completed reports-

few of the counties allowed full complete audits

Only 4 relatively small counties were noted in the recount summaries to have done full random samples as required by Ohio law.

 4 other counties were noted as having either done a random sample with demographic limits on the counties that could be chosen or

 to say they did previous random sample without recount team participation.

In the recount/audit process, there was evidence of fraud or manipulation of vote totals that were likely to have produced significant deviations

in the official vote totals compared to the voter intent.  This is in addition to failure to count or take into account undervotes due to hanging chad

as can be done according to Ohio law.  The hanging chad were a problem in getting agreement between hand counts and machine counts in many punch card counties.   

The following counties were identified as having evidence of fraud or manipulation that could have produced significant differences between official counts and voter intent

(or significant efforts to prevent recount team from assessing the validity of vote totals)

Auglaize, Ashland, Coshocton, Cuyahoga, Fairfield, Fayette, Fulton, Green, Hamilton, Hardin, Hocking,  Licking, Lucas, Medina, Mercer, Miami, Monroe, Montgomery, Perry, Putnam, Sandusky, Summit, 

Some of these counties have had followup analysis by independent analysts that provide further documentation on these counties issues, as well as large numbers of irregularity incident reports to the EIRS hotline on election day.     Some of this documentation can be found at the following 5 sites:

http://northnet.org/minstrel/alpage.htm               Mahoning and Columbus and Cleveland and etc.

http://www.freepress.org           Departments        articles

http://www.flcv.com/ussumall.html            Ohio

http://www.voteprotect.org           maps        Ohio   county

http://www.votersunite.org          Ohio

The Green recount teams in several counties also documented that the DRE touchscreens in some counties have individual votes by precinct stored in the computer memory and audits and hand recounts.   One county, Lake, produced a sample printout of 3 % of the votes from election day and carried out a handcount of this printout tabulation, which matched the computer compilation.   Which should not be surprising.   But it should also be noted that the Ohio Secretary of State forbid recounts of the touchscreen votes in other counties where it was requested.    Indications of fraud have been found in at least 3 of the other touchscreen counties in Ohio and further audit of the machines in those counties would likely have produced further evidence of these problems. The Greens are continuing their efforts in this regard, but are meeting with resistance.

            The Green recount teams in several counties also documented that the Triad Manufacturer representatives have a remote patch to the Triad compilers and can make changes to the compilers remotely.  They maintain and test them remotely, and on election day had a running tally of votes on their company website as the votes came in. 

 

 

 

Touchscreen Counties (Auglaise, Knox,  Lake, Mahoning, Franklin,

Optical Scan Counties(Allen, Ashland, Cushocton, Hancock, Hardin, Lucas, Miami, Sandusky,

Punch Card Counties(Champaign, Clermont, Cuyahoga, Delaware, Fairfield, Fayette, Fulton, Green, Guernsey, Hamilton, Henry, Hocking, Holmes, Huron,  Jefferson,  Marion, Medina, Meigs, Mercer, Monroe,  Montgomery, Morrow, Putnam, Tuscarawas, Union, Vinton, 

               

 

Allen County (Apparently optical scan) (any county can be found at following URL)    **

http://www.votecobb.org/recount/ohio_reports/counties/allen.php

3% sample not random(chosen by county)

hand count and machine count differed by one vote(marked in ink so machine would not count)

No hand count done for rest of county

No access to voter logs, provisionals, absentees until after Jan 10.

********************************************************************

Ashland County Optical Scan    *** (butterfly overvote problem)   http://www.votecobb.org/recount/ohio_reports/counties/ashland.php

3% sample not random(chosen by county)(no one seemed to know how to do random sample)

Ballots stored in open containers in staff snack room accessible to staff

{Significantly, missing from the ballot instructions is anything referring to erasures. This is important because erasures are counted as over-votes and not counted as a valid vote. There was no clarity about this on the ballot. [In other words, if you erase a mark on the ballot, the ballot automatically becomes invalid, but the instructions on the ballot make no mention of this fact.]

This year's was so confusing that there were 530 over-votes, as compared to 2000 when there were only 166 (a difference of 364 votes). This may sound confusing but bear with me, and look at both samples first. It was determined (by some estimating among the 4 witnesses present) that about 80% of the over-votes were indeed clear about the intent of the voter.}

{Confusing ballot design caused rejection of lots of votes with clear voter intent:

However, due to the odd way that the bottom group of candidates (president and vice-president) was placed on top of the write-in space, many, many people filled in both the president and the write-in while leaving the space blank [thereby invalidating their ballots].}

{Ballots filled in ink had been rejected by machine, but SOS ruled that in recount they could be counted since intent was clear}

{Absentee ballot design extremely confusing- unlikely most voters would know how to fill them out- errors likely- placement of circles by vice pres. candidate likely to produce misvotes}

{Provisional Ballots 631 total 554 counted 77 not counted

(no access to records to confirm why votes not counted- most appear due to inability to confirm registration)}

{Poll books- looked at briefly- just noted number of voters penciled in by staff on election day-

I can find no explanation for why there were none offered to us from Greene TWP 1 and Lake TWP. Further note ---- The numbers of precincts and the names of precincts did not exactly match the list on the Precinct Status Report 11/02/2004 that we were handed earlier. did not have time to look at signatures }

Some Precinct totals in the log books were strange:



PRECINCT REC'D CAST SPOIL PROV TOTAL (others not listed here matched #s)



002 1-B 600-R 354-C 3-S 6-P ? left blank by poll worker

003 1-C 520-R 291 3-S 5-P =282 possible ballot stuffing (+61)

009 2-C 369-R 450-C 3-S 6-P =360 R& C reversed?, sloppyiness? (+450)

033 GREENE TWP 2 DID NOT SEE THIS

034 LOUDONVILLE VILLAGE 2 520-R 327-C 0-S 13-P =167?

035 LOUDONVILLE VILLAGE 3 500-R 381-C 3-S 6-P =110



036 HANOVER TWP

550-R

461-C

3-S

5-P

=80

040 LAKE TWP

DID NOT SEE THIS

041 MIFFLIN 1 (+100) stuffing or err

225+50+50-R

298-C

0-S

4-P

=76+17+30

042 MIFFLIN 2

325+30+50-R

282-C

2-S

4-P

=82+11+22

043 MILTON TWP 1

400-R

319-C

8-S

3-P

=69

046 JEROMESVILLE VILLAGE

325-R

222-C

2-S

2-P

=79 (-20)

052 MONTGOMERY TWP 4

475-R

357-C

2-S

3-P

=111

053 ORANGE TWP 1

435+15-R

354-C

2-S

6-P

=82+6

055 ORANGE TWP 3

450+10+40-R

384-C

1-S

 

=88+5+15

059 SULLIVAN TWP 1

620-R

513-C

4-S

7-P

=92

060 SULLIVAN TWP 2

500-R

375-C

1-S

3-P

=120

062 HAYESVILLE TWP

300-R

219-C

0-S

1-P

=79

063 VERMILLION TWP 1

440-R

351-C

3-S

6-P

=79?

064 VERMILLION TWP 2

400-R

329-C

3-S

4-P

=61?

65 VERMILLION TWP 3 (+5)

440-R

359-C

3-S

4-P

=79

Respectfully submitted by Witness J.

***********************************************************************

Auglaize County ESE DRE   (absentees counted by optical scan)         ***  

http://www.votecobb.org/recount/ohio_reports/counties/auglaize.php

Absentees 2262   (counted by Optical Scan Tabulator)  (no random sample of election day votes allowed by order of S.O.S.)

( Often the counter and the observer would come up with different totals. If it was greater than seven they would recount. Less than seven they would call in the leader and it would be adjudicated.?)

( my judgment is that the DRE is nice to vote on but totally unreliable as a method of voting. One would hardly trust his money to a bank which had a back door which could not be locked.)

The actual cartridges from election day were then read into the tabulating PC. One cartridge was inserted twice, the tabulating software identified the error, and the Technician confirmed that the data was only transferred to the PC once. A previously known error caused the votes from one precinct to be read from the FLASH memory of the DRE rather than the cartridge. The tabulating PC would not recognize its FLASH card reader. The Technician produced his own laptop, inserted a ZIP disk, deleted the contents of the ZIP disk and transferred the contents of the FLASH to the ZIP disk. The ZIP disk was then inserted into the tabulating PC and read into the tabulation data.

Grand totals for the presidential candidates were totaled and the results were printed. The results of the count varied from the original voting reports as follows.

· The absentee ballot count was four ballots higher than the original count.

· George W. Bush gained four votes over the original count.

· John Kerry lost one vote from the original count (an additional over-vote was counted, canceling one Kerry vote.

· All other Presidential vote counts remained as originally counted.

A witness inquired as to the capability of the DRE to print an image of the ballots as cast. The witness was informed that the DRE machines in question could produce the images. The witness then requested that images represented 3 percent of the ballots cast be printed and examined. The witness was informed, by the Director of the Board of Elections, that printing and counting the images was outside the scope of the recount requirements as set forth by the Secretary of State of Ohio in his directive concerning the vote recount.

Employee alleges voting machine errors. A former employee of ES&S, ""the company that provides the voting systems in Auglaize County, had access to and used the main computer that is used to create the ballot and compile election results……a violation of county board of election protocol……Mr. Nuss was suspended and then resigned.""

********************************************************************

Brown County

random sample from hat but only used precincts with totals less than the 3% ?

3% sample matched

Poll book totals were checked against precinct totals & matched

(recounters didn't seem to understand much about vote and recount process, apparently no check of absentees, provisionals, misvotes)

*********************************************************************

Butler County (Recounters don't seem very serious- didn't give info on how sample done or details about the recount- just said it ran "smoothly".

***********************************************************************

Champaign County Wayne Rapp, V.P. of Triad Systems present Punch cards    ***        (no serious recount carried out) 

(He took the lead on how to do the recount and directed the recount)

Affidavit  

I, Witness P., want to state for the record that there were several serious problems with this recount in Champaign county on 15 Dec 2004. I believe that the recount for this county should be declared incomplete due to policies and actions by the office of the Secretary of State as detailed in the following problems.

First, I want to state that all the personnel were courteous and pleasant at all times. Further, I felt they were reasonably well informed and were doing things as their board and the Secretary of State wanted them.

There was no random selection of three percent of precincts. The precinct was chosen well before we arrived, probably days earlier because they were using printed tally sheets with the precinct Salem North printed at the top. I pointed out the importance of a random selection and explained how to do that by writing names of precincts on paper and putting them into a hat or bowl for the selection. They commented that making the selection that way they could wind up with a precinct that was too small and have to choose a second. I reiterated that I understood that could happen but random selection was important and that if they proceeded without making a random selection of precincts I would highlight that in my report. Without a truly random selection of precincts, the precincts that would be subject to a recount are easily predictable and the value of using a recount of a small portion to ensure accuracy throughout is useless. I think this is an easy concept for people to appreciate and that just about everyone connected with the BOE understands it intuitively.

The ballots were not stored in sealed containers. They were stored in latched tin containers in the same room where the counting machine is kept. The room may have been locked at night.

The signature book was not made available even though I requested it several times. Director Burden called the SOS office for clarification around 3:00 PM and still did not let me see them. She told me that I could never see it. At any rate, I needed it right away; without the signature book, I could not confirm how many people actually arrived to vote.

I was not allowed to see the absentee ballot envelopes that may have arrived late or had some other problem with them. I was given only a verbal count of how many absentee ballots were sent out and how many were returned.

I was not allowed to see any provisional ballot that was rejected. I was not allowed to see a summary of rejected provisional ballots.

In every case that I could not see documents that were necessary, I was told that I could see them after January 10, 2005, per orders of the Secretary of State. I believe the Champaign county BOE offices were orderly and everything seemed to be proper, to the credit of Dir. Burden and Dep. Dir. Nash.

Witness M. is also willing to sign an identical affidavit.

******************************************************************************

Clermont County (Punch card)      ***

http://www.votecobb.org/recount/ohio_reports/counties/clermont.php

3% random sample was not random; only small precincts included; hand count didn't match machine but didn't follow Ohio rule to count all by hand There was indication of inconsistent policies in counting favoring Bush and evidence of tampering according to witness. County officials uncooperative; would not let observers see info needed to confirm results, such as overvotes.    Kerry votes on some cards had stickers over them so could not be counted by compiler.  No explanation given.         http://rawstory.com/news/2005/index.php?p=7

***********************************

Clinton County

only one observer allowed other than Repub observer. Recount was very cursory and quick.

witness overwhelmed and intimidated by process and didn't do much checking. Didn't notice any glaring problems. No discussion of whether 3% sample was random. Didn't check absentees; told provisionals mixed into regular ballots.

*************************************************************************

Columbiana County

No discussion of whether 3 precincts handcounted were randomly chosen;

056 Knox Twp NE 46 undervotes or overvotes out of 599

029 Salem 4-C 33 undervotes or overvotes out of 422

015 E. Palestine 3-A 63 undervotes or overvotes out of 612

143 provisional ballots rejected

only the 3 sample precincts were looked into by witnesses.

************************************************************************

Coshocton County Optical Scan       ***    http://www.votecobb.org/recount/ohio_reports/counties/coshocton.php


The county voluntarily opted for a full hand recount, likely for reasons related to the large discrepiency found between the official certified numbers and the results of the recount, with it being clear that there had been ballot counting for unofficial reasons between the certification and recount time.

Once again, ballot security is an issue. Boxes were open prior to our arrival. They are kept unsealed in a locked room at the Board of Elections. Vote totals were distributed to witnesses with the apparent intention of misleading them. The hand count totals were compared to totals furnished by the BOE dated/generated(?) 12/10/2004 as part of some unknown "unofficial" process rather than the official certified SOS totals. These totals were incorporated into a three-page list indicating vote totals precinct by precinct. These vote totals were distributed to witnesses in a manner that suggested they were the totals certified by Secretary of State Kenneth Blackwell, on December 6, 2004. They were in fact not the totals certified by Secretary of State Blackwell. The lists provided to witnesses are noted to be "Official Nov. 2, 2004" and are dated 12/10/2004 10:25:28am. These include pages 1, 2, 3, of 96. There were major differences in these totals from the certified totals.

Some issues were raised during the hand count that warrant mention. The failure of the board to allow witness presence prior to the handling of the ballots. BOE personnel (employee/member) alluded to the counting of some precincts prior to the start of the recount. Witness, Patricia Stout, stated that "one of the precincts I observed was already separated into Republicans, Democrats and Other. It was clear that this pile had been counted." The count was completed and totals reflected a net loss of 13 votes for George W. Bush, and a net gain of 34 votes for John Kerry. The hand count totals were compared to totals furnished by the BOE dated/generated 12/10/04.

Signature book examination conducted by the witnesses for the Democratic candidate looked at two different precincts, with BOE personnel recounting Jefferson Twp. Precinct. Discrepancy appeared to indicate a pattern of a 7-8% loss of votes for the Democratic candidate.

*********************************************************

Cuyahoga County   punch cards   ***

The 3% sample was not done in a random manner. Precinct selection was done on the basis of only choosing precincts with 550 or more votes, and a cross-section of areas ---- one East side, one West side, one affluent, one non-affluent.

Only ballots with holes punched through, with 2 or less chads remaining would be counted. Dimpled chads would not be counted, even if the entire ballot was dimpled chads.

there will be no visual inspection of the remaining 600,000-some ballots in the county.

They are not counting the rejected provisional ballots and uncounted absentee ballots even if upon inspection by the witnesses errors are found and can be demonstrated. Vu says the City Prosecutor says the approach of not adding these votes into the recount totals is fine.

Witnesses were then able to look at the signature books of 60 different precincts, including those of the 3% selected and 26 other precincts of concern. They were looking to make sure that the number of signatures in a precinct matched the number of ballots cast in that precinct. In many cases they did not. From what they were able to get through, witnesses found that signature counts were very much different from the official recorded number of ballots.

Anomalies were found. Almost all of the witnesses that I spoke with felt that the ballots were not in random order, that they had been previously sorted. There would be long runs of votes for only one candidate and then long runs for another, which seemed statistically improbable to most.

the alphabetical log books were taken away from witnesses, and replaced with only the log books of the 3% hand-recounted precincts, in addition to about 30 of the specifically-requested precincts. The volunteers began examining these log books.

In at least one instance, the rotation (meaning the order the candidates are listed) in the booklet did not match that of the actual ballot. In another instance, the page listing the Presidential candidates was missing completely from the booklet. An examination of both the transfer case and the black grip did not turn up the missing page.

Attachment: Legal Authorities

(We have full copies of all references and can fax them as needed).

Full visual inspection

a) Secretary of State (SOS)2001-10 Directive 4(Recount)(a): Ballot cards must be inspected for hanging chad.

b) SOS Conference Call with Statewide BOEs of 12/8/04: Question 7 asks whether the language in subsection (a) above discussing the inspection of each ballot card refers to the 3% manual count or all ballots. The answer the SOS gives is: All ballots must be inspected."

c) SOS Conference Call with Statewide BOEs of 12/8/04 at question 8: "Is ballot inspection a required step of the recount?" Answer: Yes. We have copies of all SOS conference call minutes to which we refer. Can fax them over.

Precinct Requests: Concerns whether parties to the recount have a right to request hand count of specific precincts within the 3% hand out.

a) Cuyahoga County Board of Elections (CCBOE): Recount: Standard Operating Procedures Guide: F4: "Any requests for specific precincts must be made, in writing, 24 hours before the recount procedures begin."

b) Guide clearly contemplates submission by parties of specific precinct requests (thought BOE is not required to accept them all).

c) Greens, Dems and Libertarians all submitted specific precinct requests yesterday, well before 24[-hour] period.

Rejected provisionals and uncounted absentees: recount includes not just observation but correction of rejections as warranted:

a) Ohio Statute: R.C. [Revised Code] 3505.27: Counting of Votes: "Unless otherwise ordered by the secretary of state or the board of elections, the counting and tallying of ballots shall be conducted according to procedures prescribed by the board of elections that assure an accurate count of all votes cast . . . . The board shall prescribe additional procedures as necessary to assure an accurate count of all votes cast."

b) Section should apply to counting of votes at any point during election process, initial tallies or recounts.

c) Vote count will not be accurate if recount does not include revision of inaccurate rejected provisionals and uncounted absentees. Many challenges of inaccuracies here will come from witnesses.

Precincts for hand count:

a) SOS Directive 2001-10 requires under 4(Recount)(d): "The board must randomly select whole precincts whose total equals at least 3% of the total vote."

b) See also Cuyahoga County Board of Elections (CCBOE): Recount: Standard Operating Procedures Guide F3 (3% recount must be "randomly selected").

c) This entails a scientifically random sample, including precincts of varying sizes.

d) Statement on selecting a "random" sample by Professor Cyrus Taylor, Case Western Reserve University (CWRU): There are at least three problems with the rule that precincts will be chosen for a hand count from those in which at least 550 ballots were cast:

(1) The small fraction of precincts (only 8% of all precincts qualify for their ballots being closely examined in the hand recount) are not reflective of the county as a whole. Overall, the qualifying precincts have significantly higher proportion of votes for Bush than for Kerry, yet the county voted for Kerry by a 2:1 margin.

(2) Precincts that have significant problems on Election Day are that much less likely to reach the 550 minimum, and will thus not be rechecked in the recount.

(3) The vast majority of precincts have less than 550 ballots cast. It is thus possible in principle to fiddle with the returns from most of the county without fear that they would be checked in the hand recount of 3% of the returns.



additional independent notes:

CUYAHOGA COUNTY - (1) ""more than 10,000 voters"" were kept from voting; (2) in a mostly black precinct, confused results for Kerry and third party candidates indicated ""fraud, error, or mistake""; (3)""there was an effective denial of the right to cast a provisional ballot and have that provisional ballot counted. 8,099 were ""ruled invalid,"" about 1/3 of those cast; (4) ""voters were misled when they received phone calls incorrectly informing them that their polling place had been changed""; and (5) in some cases, ""arrows on the absentee ballots did not align with the correct punch hole"" which ""led to voters casting a vote for a candidate other than the candidate they intended to support.""

*********************************************************

Defiance County

Coordinator did not indicate whether 3% sample was random. Apparently asked no questions and did not look at records. Simply stated that the hand count matched the machine count, so rest were counted by machine uneventfully.

*****************************************************

Delaware County punch cards

No indication as to whether 3% sample was random. Hand counted 4 precincts.

ES&S contractor participated in the machine counts, kept machine running, reset when needed, etc. The hand counts matched the machine counts, and the rest were run through the machine.

*************************************************

Fairfield County Punch cards    ***

The 3% sample was not random. the Democratic and Republican board member each selected a precinct to recount. They were Lancaster City 2C and Pickerington C. Pickerington C did not match. 4 votes difference. 2 tabulators used and both failed often. One could not be used.

Recounted and again didn't match. After agreeing that the entire county would be hand counted, they called the SOS and we were told a decision had been made to do another machine count the next day using a new machine to be brought in by the vendor. They also indicated that the recount would be conducted under much more restrictive conditions on the witnesses.

The BOE was informed by the Greens that this was a violation of the recount law.

*********************************************************

Fayette County punch cards   ***                    

The 3% sample was not randomly selected. 2 out of the 38 precincts had been chosen: one urban (Washington 2A) and one rural (Wayne East). They totaled just over 3% of votes cast.

The undervotes (no presidential candidate chosen) were almost all punched in an odd-numbered punch hole. The overvotes punched more than one even-numbered punch hole.

The recount for Washington 2A matched exactly, including 3 undervotes, and 1 overvote.

The recount included 15 absentee ballots, 7 provisional ballots. All had been put into the metal box for the precinct. The recount for Wayne East matched exactly, including 2 undervotes and 4 overvotes. It included 9 absentee ballots and 3 provisionals.

The Democratic volunteer and I looked at the rejected provisional ballots. They were bound by rubber band and each was in an envelope with a reason for rejection handwritten on it. We were given a typed list (attached), alphabetized by last name, of rejected provisional ballots. There were 46 names on the list. 38 voted in the wrong precinct (WP) 8 were not registered (NR)               (83% of rejected provisionals were for voting in wrong precinct)

(undervotes and overvotes appear to be due to unreliablility of the voting system)  

FRANKLIN COUNTY    (  )     no green recount report filed to date

- (1) In some absentee ballots, there was lack of alignment in the punch card hole, such as occurred in Cuyahoga County; (2) Widespread dirty tricks in minority precincts supplying incorrect poll location by fliers and calls by someone ""claiming to be from the County Board of Elections..""; (3) ""there was a discriminatory assignment of more voting machines per registered voter to precincts with more white voters than African-American voters.."" with resulting very long lines. ; and (4) there were ""numerous reported instances of vote hopping (in which a voter selecting Kerry for President saw the choice displayed on the machine ''hop'' to Bush for President  or was unable to vote for Kerry).""

 

Fulton County (Punch Card,  Triad tabulators)          ***

No discussion of whether 3% sample was random. The Director for Fulton told me that Triad is able to reprogram the computer to count only the Presidential ballots by remote dial-up. I do not know when after the election that occurred.

The hand count matched the machine count and the rest were run by machine.

**********************************************************************

Gallia County    **

3% sample was random sample. Hand count did not match machine count(differed by 1)

Quit to decide whether to hand count whole county as per Ohio rule.

*****************************************************************

Green County Punch cards Machine count performed by Triad President Brett Rapp     ***    (no serious recount allowed)( misvote irregularities)

3% sample not random. 3 precincts chosen. Witnesses not allowed to closely witness or monitor counting, only from across the room. Witnesses were unable to look for hanging chad.

Only one of the 3 precincts had the hand count match the machine count. The hand count was repeated until it matched the machine count. Took one more time for one precinct and 3 more counts to match for the other.

We were not allowed to see the poll books or signatures, but were told we could return the next day to view the poll books of the 3 precincts chosen for hand count.

Information for four rejected provisional ballots was provided for precincts in recount only for Precinct 442. There were 3 absentee ballots rejected for being received after November 2 in Prec. 442.

The precinct won by Bush(31) had no undervotes or overvotes. The 2 precincts won by Kerry had large numbers of misvotes. Precinct 224 had 15 undervotes and 1 overvote. Precinct 442 had 10 undervotes and 1 overvote. Overall the County had 992 undervotes and 332 overvotes.

Overall Bush gained a net vote of 10 votes in the recount.

*************************************************************************

Guernsey County Punch cards (Triad tabulator) (Triad rep present)   ***

3% sample was not random. 1 precinct was hand counted and matched machine count. The Triad rep had tested and prepared the machines prior to our arrival.

Ballot storage appeared to be secure.

The log book signature numbers also matched the total number voting for this precinct, but only looked at one precinct.

Hamilton County   Punch cards                                                 ***     voter logs not compared to official votes;

Sample:  it was explained to us by John Williams (Director of the Hamilton County BOE) that 3% of the Hamilton vote was equivalent to about 30 precincts. He indicated that they had selected these precincts by taking every 34th precinct beginning with 1A. The BOE believed that the final list was representative of both city precincts and those in outlying areas.     While the handcounts did not exactly match the machine counts for all precincts, difference was explainable on further investigation as overvote effects.  

3400 provisional ballots were rejected,  the reasons were not fully determined.    But  "Approximately 400 of those [Hamilton County provisional ballots], he thought, were rejected because they were cast in the wrong precinct, despite the fact that they were cast at the right polling station (i.e., at a polling station with more than one precinct)."  (ie, there was malfeasance on the part of poll officials and the votes were from legal voters and should obviously have been counted)

Hancock County     

No indication of whether 3% sample was random, witness did not appear to be knowledgable about voting systems,  no discrepiencies noted,

No indication that vote logs or absentees or provisionals were checked.

Hardin County   (Optical Scan)   ***

The sample precinct was not chosen at random, but was was pre-selected."

"A total of seven Diebold tabulators were used and a test run was done on three of them. We requested that the other four machines have a test run done and this was denied."

"In the morning, Brandon [the Diebold technician] informed K that he could not stand behind the tables holding the tabulators while the recount was being conducted because this was a "security issue" as the discs with the election results were behind the tables."

"Another obsever, Cathy from the Democrats, reported she had seen Brandon the day before alone with the tabulators testing the machines."

"We found seven ballots that were cast in the correct polling station, but incorrect precinct and were disqualified."   (Poll worker error)

— Green Party Observer     Hardin County Recount

 

Henry County      **

No indication of whether sample was random.     Hand count did not appear to match machine count(difference of 4 votes) but may have been due to different treatment of overvotes by the 2 methods so rest of county was machine counted as if there was a match. 

Hocking County (Punch Cards,   Triad tabulator) ***

Sample was not random, observers objected.   BOE was not cooperative regarding doing a serious recount.   Manufacturer rep was told the number Counted by hand count prior to doing the machine count. 

David Cobb Reveals Reports of Tampering with Voting Machines

Dec. 13, 2004

David Cobb: "Representative Conyers, I have the most recent breaking information. It will be very brief, but this is, I think, very important. It's probably the most troubling incident that has been related to my campaign. It was related only on Sunday December 12th, regarding the very shocking event that occurred last Friday, December 10th."

"The representative from Triad Systems came into this county's board of elections office un-announced (that is, on this Friday). He said he was just stopping by to see if they had any questions about the up-coming recount. He then headed into the back room where the Triad-supplied Tabulator (that is, the machine that COUNTS the ballots) is kept."

"This Triad representative told them there was a problem with the system, that the system had a bad battery and had — quote — 'lost all its data' — end quote."

"He then took the computer apart and started swapping parts in and out of it and another 'spare' tower PC also in the room. [He may have had spare parts in his coat as one of the BOE people moved it and remarked as to how very heavy it was. He finally re-assembled everything and said it was working but to not turn it off.]"

"He then asked which precinct would be counted for the 3% recount test, and the one which had been selected as it had the right number of votes, was relayed to him. He then went back and did something else to the Tabulator."

"The Triad Systems representative suggested that since the hand count had to match the machine count exactly, and since it would be hard to memorize the several numbers which would be needed to get the count to come out exactly right, that they should post this series of numbers on the wall where they would not be noticed by observers. He suggested making them look like employee information or something similar."

"The people doing the hand count could then, he said, just report those numbers no matter what they actually counted in the ballots. This would then — quote — 'match' — the Tabulator report for this precinct exactly. [The numbers were apparently the final certified counts for the selected precinct.]"

Holmes County    (Punch Cards,  Triad Tabulator)   **

Precinct ballots were not sealed after the election.    Recount observers were not allowed to look at “spoiled ballots” that were rejected.  No count or breakout of rejected votes or misvotes given.

No definitive statement of whether the sample was random, but appears sample was not random. 

Huron County   (Punch Cards,  Triad Tabulator)  **

Sample(3 precincts) was not randomly chosen.     The tabulator was tested by the manufacturer the previous day without a Recount Witness present.

The first precinct was counted and all election workers came up with different totals.   A representative of the SOS then took over and helped with another count to develop a consensus count.   The machine tabulator jammed about five or six times each time a precinct was counted. Each time it jammed, the machine had to be reset. The totals for precincts #1 and #3 were off by one vote.  It's worth pointing out that precinct #1 was hand-counted three times. Precinct #3 was hand counted two times. Each precinct was machine counted two times.  The discrepiences appeared to be caused by 2 cards that were stuck together in one precinct and a hanging chad in the other.    The other precincts were then counted by machine.

Time was short so a copy of voter logs for 2 precincts was made with SS#s blacked out, to be looked at later.  No analysis of absentees or rejected provsionals was made.

Jefferson County    (Punch Cards,  Triad Tabulator)  **

Appears they may have done a  random sample of precincts that were approx. 3 percent of total votes.  But the precincts were preselected without recount observer participation.    Hand count differed from machine count by 1 vote.     Poll workers separated by undervotes, overvotes, and o.k. votes on the presidential column. On several occasions, hanging chads were knocked clear by poll workers or their supervisors. The actual undervotes for the county decreased by 5 ballots, based on the recount computer tabulation.    It is not stated whether a decision was made to do manual recount of all precincts, but apparently not.

Knox County  ( DRE, Microvote; Absentees and Provisionals by Optical Scan)

The Microvote DRE has a permanent memory chip and can print of paper tape. 

·                     Full machine count of all absentee and provisional ballots (4309) then performed, with ballots cast tabulated at 4281. Board of Elections employee stated that the discrepancy was within the margin of error.    There apparently were undervotes.

·                     Tabulation of cartridges experienced computer malfunction during data reading optical scan ballots. Computer would freeze, necessitating rebooting and rechecking ballot count.

·                     Cartridges were then read and all optical scan and cartridge data combined for a summary report.

·                     Permanent paper audit trails inside machine not available to witnesses, available audit trails matched within margin of error.

Poll and signature analysis for some precincts was to be carried out.

Lake County     (DRE,     )            **

The DREs have voter choices in memory and a sample of machines was chosen and a handcount of the memory printout carried out for the chosen machines.     The handcounts and machine counts matched. 

There were many write-ins for both Bush and Kerry which could not be counted because they were not write-in candidates.

The wording on the law must be changed to allow write in votes for any qualified candidate, not just those who are qualified write-in candidates. Clear voter intention is being disregarded over a technicality.   ?

No indication that absentee votes or provisional votes were looked at, or  voter logs.

******************************************************************************************************************************************************************************

Licking County     (    )     ***

 

Our team, to be frank, was greeted on opening day, Wednesday, by a rude and arrogant election board and senior staff which refused to discuss its recount procedures or respond to questions, let alone challenges.

As you may have read in the Advocate, one of our group was even threatened with arrest for challenging board dictates which she found contrary to law."

"The board, seemingly informally without a recorded vote, decided we will not be able to inspect the voting machines and their ballot assemblies to insure correct rotation of candidate names."

"Neither will we be able to check rejected provisional ballots to determine if denials were justified."

"Nor will we have any say as to which precincts will be used in the initial three percent sample count, precincts supposed to be selected 'at random.'"

"At one point, the elections director even told us not to slow the ballot inspection by pausing to look at "overvote" or "undervote" ballots, a ruling thereafter reversed by the elections board chairman."

We were told, for example, that it is not board policy to attempt to determine voter intent on ballots "dimpled" by a weak punch which leaves only a mark or indention. The vote will only count if the stylus cleanly separates the chad from the ballot.

"[regarding the undervotes] ... witnesses were surprised to see how many Licking County voters seemingly braved storms and crowds to go to the polls, stand in long lines and then NOT vote for president ..."

"During two days of observing ballots from 26 precincts, I found nearly 200 such cases of undervotes or overvotes, many of them involving absentee ballots."

"Something strange is going on here, and we're getting similar reports from other counties."

 

— Green Party Observer    Licking County Recount

No report on the outcome of the sample recounts.

Lorain County    **

Our requests for guidelines for the recount were not accepted by the BOE. It was made very clear that it was their recount, with their rules, and we could only observe. The list from McTigue was filed by the board but not used. Bad beginning feelings.

"The three percent was chosen by random by the BOE but not witnessed by us. This made us feel that the BOE had something to hide."

"Ballot counting BOE personnel went very fast with the recount. It was difficult to observe everything. BOE personnel were instructed not to speak with witnesses."

"We did get to see rotation pages and found no problems. Should be noted that employees of BOE had access to these before and after election. Not locked."

"We did get to see precinct boxes ... Should be noted that employees of BOE had access to these before and after election. Not locked."

"Blackwell's 'correct precinct' ruling really affected this election. Provisional ballots are rejected with a limited amount of knowledge, without using inferences favorable to the voter, and a fair appeal procedure is not in place. WE NEED TO GET THIS CHANGED BEFORE 2006!"

We saw the overvotes and undervotes of the three percent. We asked to see the remaining 97 percent of overvotes and undervotes, but were not allowed. As a result, we were unable to figure out whether there were any patterns to these uncounted ballots.     No absentee ballots were seen.  Green Party Observer              Lorain County Recount

Lucas County   ( Optical Scan, Diebold)   **

The ballots for the preselected precincts were first counted by hand and then counted by the optical scanners"   "The problem: I witnessed 3 ballots physically altered."

"The apparent reason for this was to ensure that the vote count produced by the optical scan machine matched the hand count, and thus avoid hand-counting all of the ballots in the county."

"These two ballots were physically changed by the election officials... This was done before the ballots were fed into the optical scanner. I asked one of the election officials if the ballot in case 2 would have been recorded as an under-vote on election day. She first said yes, then she said she did not know..."

— Green Party Observer

"It has also been reported that the observers were not allowed close enough to be able to see the numbers used for the test run or the results from the counting machine of that test run."

"They were able to see the test being run but not close enough to see the numbers on the sheet that provided the preliminary test numbers, and then the board of elections officials actually left the room with the results from the machine count, which again the observers couldn't see, and the preliminary numbers then returned and reported that it had come out OK."

"The group then was given a break for lunch and upon their return were kept waiting for a while. It was explained to them that they were waiting for the technician from Diebold to reprogram the machine."

       Green Party Observer  Lucas County Recount

LUCAS COUNTY -- (1) there was ""discriminatory assignment of voting machines to precincts; and (2) machine errors ""snarled the process throughout the day. Jammed or inoperable voting machines were reported throughout the city. Lucas County Election Director Paula Hicks-Hudson said the Diebold optical scan machines jammed during testing in the weeks before the election.""

Mahoning County   (DRE, ES&S)  (Absentees and Provisionals by Optical Scan)     ***                     (recount of absentees & provisionals only)

"Gloriane Leek, witness for the Democrats, ... specifically asked to see provisional ballots that were not counted because the person was in the wrong precinct but at the right poll."        "They let her look at the ballots and she estimates that 50 percent of the provisionals filled out for Mahoning County were not counted because the person was in the wrong precinct but right poll place."    "Gloriane Leek asked that those 50 percent be recounted and put in the recount certification. The board decided no."   "They said it is Ohio state law under the 6th District Court and put forth by the Secretary of State that provisional ballots cast in the wrong precinct are not allowed to be counted and therefore would remain uncounted."

Gloriane Leek also brought up the issue of absentee ballots that were rejected. However, those ballots were not brought out and were not counted in this recount.

Gloriane Leek also raised the issue of distribution of voting machines to precincts. She said precincts in lower income neighborhoods had less machines and not enough machines and therefore longer lines and more voter disenfranchisement.  She said the Democrats might file suit under the 14th amendment to deal with this issue.

As a sample test for the recount, 600 absentee ballots were selected at random to make up 3 percent. Those 600 were counted by hand and then put through the optical scanner to see if it matched up. For Kerry, there were 7 more votes during the hand count than during the machine count.

At the end of the optical scan machine count, the over/under votes were discarded and the ones in the hand count pile were taken and counted by a woman named Joyce. We all watched. For some of the votes, she could not decide because of stray marks or lightly colored in circles so she made a new pile specifically for the board to look at. She put these votes in an open basket. It was decided that the board would have to look at these votes and determine how they should be counted if at all.

At the end of the optical scan machine count, the over/under votes were discarded and the ones in the hand count pile were taken and counted by a woman named Joyce. We all watched. For some of the votes, she could not decide because of stray marks or lightly colored in circles so she made a new pile specifically for the board to look at. She put these votes in an open basket. It was decided that the board would have to look at these votes and determine how they should be counted if at all.   When the board reconvened, they handed us spread sheets with the recount numbers on it. Kerry had 109 more votes and Bush had 47. These are vote increases. The board then certified the results and the observers left.

I, along with M., stayed behind and asked what about the votes in the basket marked "must be looked at by the board." They were never looked at. Mike Sciortino the director and one other board member took M. and myself to the room that holds the votes and together we looked at the basket with the ballots that was marked to be looked at by the board. Neither M. nor myself ever touched the ballots.

I asked what their plan was to deal with these ballots that were not counted even though the recount was certified. They said tomorrow the board would reconvene and look at the 23 ballots and redo the results of the recount and re-give us all spreadsheets.

My concern is that for about 2 hours this box was out of the observers site and I do not know for sure that 23 was the number of ballots originally in the box. My concern also is that I was not given a time that the board would look at these ballots and these are highly contested ballots and I think the board might look at them without observers there. Also it is of great concern that if I had not said something these ballots would have been lost.

"...It should be noted that during the recount and the board meeting one board member was missing the whole time. His name is Robert J. Wasko, a Democrat, who was away on a trip and therefore the board was uneven with 2 Republicans and only one Democrat for the whole of the recount."

— Green Party Observer    Mahoning County Recount

 

Marion County (Punch Cards, Triad Tabulator)

They created a 'zero count deck' (test deck) and ran it before we arrived."

"They also had four precincts already 'chosen at random'"

"... Although this was not ideal, after spending the day with the people in this office, I have no reason to suspect that there was a 'hidden agenda'..."

"All poll books and logs were laid out for [us] to look at and additionally there was a person who was willing to answer all questions regarding the poll books."

"If I were grading the Board of Elections of Marion County I would give them an A+."

— Green Party Observer

 

Medina County   (  Punch Cards, Triad Tabulator          )   ***

The precincts to recount were not chosen randomly. According to Mr. Thomas E. Wolfe, Chairman of the Medina County Board of Elections, 'I took two Medina precincts because I knew they had a high vote. I could have taken five precincts, but then we'd be counting for another hour.'"

"...Witnesses were not allowed to review the selection and preparation of the ballots to test the voting machines, nor did we have an opportunity to examine them, either before or after they were used to test the machines. We did not see the testing of the machines..."

"Two ballots in Medina City 4F had votes in both position 8 and position 9. Those ballots were not rejected as overvotes, but were counted as votes for George Bush (position 8). I was refused time to examine these ballots in order to ascertain positions of other votes, and these ballots were not saved for Mr. Wolfe to examine..."

According to Mr. Wolfe, in Medina City 3F, "Two girls with styrofoam and punchers were sent out to the line because the lines were so long." Some votes were offset from the correct voting positions. In Medina City 3F, a ballot with a vote on 11 that did not correspond to any candidate's position also showed every other vote to be offset by one, thus making the intent of the voter clear, but this ballot was rejected according to Al Junke. (Kerry vote)

BOE Director Susan Strasser, BOE Chair Thomas E. Wolfe, and NE Ohio Field Representative for the Secretary of State, Madhu Singh all refused to allow further inspection of the ballots from Medina City that I requested to view in more detail. They stated that our purpose was to decide how the computer would count the vote, and not to determine the intent of the voter. ( ?)

"Tom Wolfe, Democratic Board Member, directed the recount. It was apparent to everyone in the room that Tom Wolfe and the other board members were more concerned about getting the 3% hand recount done without any change in the tally [any change would allow for 100% hand recount of the County] than in having a fair investigation of the vote in Medina County."

"We are disappointed in how Tom Wolfe selected the precincts for the 3% hand count. He was aware of several "problem" precincts, that we had identified, but instead selected two precincts in Medina that were part of the Medina School Levy recount last Monday. That meant that those ballots had been out of the double lock situation and were handled several times since Monday."

"We believe that Tom Wolfe should have selected precincts that were not subjected to such handling. We also believe that Tom Wolfe should have selected precincts that showed anomalies. We identified one precinct that had 37 ballots without a recorded vote for president [6.2%]. The ones he selected had about 1% "no votes."

"They were very secretive and stonewalled when we asked to look at anything beyond the ballots. Even though we asked, we were not permitted to see [for each of the 3 precincts being recounted]: the voter signature books, provisional ballot tallies, provisional ballots, the actual machines and ballot booklets used at the precincts. And they did not let us see the testing of the tabulation machine."

"Other problems with the voting process in Medina: In Medina City 1B, where there were long lines, the police were standing outside all day, walking up and down the line, looking at people, and looking at cars according to Al Junke."

— Green Party Observer
Medina County
Recount

******************************************************************************************************8

Meigs County   (Punch Cards, Triad Tabulator)  ***                       (sample apparently not random & handcount didn’t match machine count)

10 precincts written on paper were put in a can and John ... pulled the paper with the precinct for the three percent hand-count."

[How the 'random sample' of 10 precincts written on paper were selected is not specified]

"During the hand-count, Ms. Frymayer [Dep. Director] asked Mr. Rapp [VP of Triad Systems] about a '5' that she had and he stated that 'he would show her what to do with it.'"

"I realized that the hand-count did not really tally ... I objected and stated that ... the hand count was still wrong. Ms. Frymyer objected and in so many words stated that ... a hand-count was not necessary."

"Please note that as the three percent ballots were starting to be tallied, Mr. Rapp handed me his card, as well as Ms. Spence [Southeast Ohio Regional Representative for the Ohio Secretary of State]. At this point, I realized what I was up against and why no one was up-front about who they were."

"When I learned he was in the room, as the recount was being started, I became extremely nervous and I did not feel like I could ask Mr. Rapp to leave the room."

"It felt like all that was happening was way over my head, and at the time that my role as a witness was not to interfere with the recount, no matter how it was done."

"At about 3:30 pm I realized that the cross-reference of the signatures in the poll books had to be done ... We reached an agreement that I could come in on Monday... I don't know if I want to go to the BOE in Meigs and I think maybe someone who is more confident should go."

— Green Party Observer

 

Mercer County( Punch Card, Triad tabulators)   ***

"... the three percent of ballots to be recounted were already selected in advance [i.e., not random]"

"[We were] forced to stand up during the entire recount process, which seemed to be a needless inconvenience that could easily have been resolved."

"... we were denied access to the poll registry ... "

"I asked our local contact person if Nader's name had been covered on the actual punch card units at the polls and she replied that they had not been and that this had been controversial."

"I specifically asked [the county's attorney] if it was possible for us to get a number on [the Nader votes, and was told] ... the only way we could get this number would be to request and pay for a complete hand recount because the company that provides the tabulating equipment and computer software had come in before the election and reprogrammed everything to simply ignore the Nader votes."

"I asked who the vendor was, and he replied Triad Systems ... [and] that, as a matter of fact, they [Triad Systems] had been at the Mercer County Board of Elections the day prior to the recount. He further stated that he was told that there was some problem with the tabulator, and that the technician had replaced a switch."

"... the technician who had done the work was still on the premises ... It turned out he had actually disassembled the tabulator at 7:30 that morning. He said he'd replaced a box containing a switch, and then made a remark to the effect that he didn't have anything to do with the software, that he's just a hardware technician."

"Unlike other counties at which the Wisconsin volunteers were present, our volunteers did not have the opportunity to see what happened when the data was transferred from the tabulator machine to the computer compiling the votes. The officials present did claim, however, that this computer is not networked to any other."

"I would not accept this on its face since they were not honest about the equipment being serviced the day of the recount. Surely they all knew this had occurred."

— Green Party Observer

 

Miami County  (Optical Scan, ES&S)  ***  (no random sample, preselected sample, hand count vs machine did not match but did not follow rules & do a full hand recount)

The morning began in an extremely structured and intimidating format, with sworn oaths and other formalities."

"Someone ... from the Miami County Prosecutor's office instructed the observers that they were not allowed to make challenges and that they were present only to watch, not verify. We were also informed that it was not necessary to be able to see the votes on the ballots, only to view the process."

A test deck was not supplied by the Board of Elections, but rather was created by witnesses.  In fact, before the witnesses insisted on a test deck, the B.O.E. was not planning to run any test deck at all. Note of irregularity: ballots seemed to already be sorted by candidate — how would this occur?

"Precincts chosen for the three percent hand count (pre-selected, not randomly)"

"There was a discrepancy between the hand count tally of the sample three percent ballots and the computer print-out tally of the optical scan of the ballots."

"... The discrepancy should have been grounds for a complete hand recount of the entire county, based upon Provisions for Recount, Ohio Revised Codes 3515, Section F5.f."

"However, a representative from the Secretary of State's Regional Office by the name of Jake Fashner interpreted the code [otherwise]."

"The Board of Electors agreed with Mr. Fashner to deny our request for a full hand recount and moved to accept and certify the machine recount as the official results of the Miami County recount."

"... Again the Board chair asked what we wanted and I said that because there was a discrepancy a full recount was in order. He went into the staff office, perhaps to check the website. Jake Fashner directed comments toward us to the effect of "Look, there will be no hand recount."

"Mr. Fashner and the Chairman of the Board suggested if we have further concerns we might take them up with ES&S."

"... a significant number of spoiled ballots (about 500) that could not be read by the optical scan machine had to be remade by B.O.E. staff with the intent of the voter being replicated to the best of the staff member's ability. It should be noted that this is a subjective process and that in the small sample we reviewed, there were questionable choices made previously during the initial counting by B.O.E. staff members in their attempt to transfer the voter's intent to a new ballot."

Director Steve Quillen agreed to email provisional ballot information to J. and A., which has not yet been supplied

— Green Party Observer     Miami County Recount

(Recount team apparently did not check voter logs against official totals or look at provisionals, undervotes, overvotes.)

Monroe County  ( Punch Cards, Triad Tabulator  )           ***      no random sample,  hand count did not match machine count, process suspended

  "Precinct 11 was pre-chosen to be counted. The reason for choosing this precinct was due to its being only 4 votes above the 3% number..."

Test Deck: Mr. Rodefer has no recollection of any test deck being run before the recount began or prior to running the 3% hand-counted ballots.

There were several machine runs to try to produce a match. "No two runs were alike and no run equaled the hand count, however the machine did operate and produce a count each time according to Mr. Rodefer..."

"They believed the problem was in the machine at this point; Triad called, to send repair person immediately (4 or more hours to arrive) to fix or bring new machine."

"BOE [Board of Elections] did not proceed to full hand count, but voted to suspend the recount and 'continue meeting' tomorrow (Wednesday 12/15) at 9:30 AM with fixed machine or new machine from Triad."

Suspended recount process. No more information available.

— Green Party Observer    Monroe County Recount

 

Montgomery County   (Punch Cards, Triad Tabulator  )   

"It seemed that the process was less an attempt to see if the machines could accurately read the punchcards and more of an exercise in ensuring artificially that the hand recount and machine tabulation of the same three percent sample were a match."

"Hanging chads, which caused so many problems in Florida in 2000 and which still account for varying totals from election night to the recount, were physically and forcefully removed from ballots once the Board of Elections directors had examined and discussed a questioned ballot to determine the voter's chosen candidate."

"Of course the hand recount will match the computer tabulation of the same three percent sample if the ballots are prepped to ensure that the machine cannot possibly misread the chosen sample."

"If the change in the three percent sample is considered statistically relevant and is extended to the full 100% of the vote, it would result in a pickup of greater than 250 votes for Kerry."

"The staff was willing and able to show us anything we requested and provide us with printouts of all critical data. We even were able to receive the list of rejected provisional ballots with each individual's personal information."  in Montgomery County, 1,881 voters cast votes that were never counted. 767 of these rejected provisional ballots were cast by duly registered voters who happened to be in the wrong precinct.

"[The] "undervotes" for which a candidate was designated and voter intent was clear heavily favored Kerry."    

"In some precincts that ran out of envelopes (which bear a code distinguishing the correct precinct), envelopes were borrowed from neighboring precincts."

"While the envelope may not have been correct for a given voter, in certain cases the ballot itself was for the correct precinct and as such the vote was counted."

Additional notes: Despite clearly distinguishable voter intent, pursuant to Ohio law, "pregnant" or "dimpled" chads were not counted as votes for a candidate. These "undervotes" for which a candidate was designated and voter intent was clear heavily favored Kerry. In addition, one ballot was discovered which had been placed in the cardholder backwards. As a result, the punches (which did go entirely through the card) were from back to front and didn't line up with the chads. Although voter intent could have been determined by overlaying the reversed ballot with a new ballot and punching through the pre-existing holes, this was not done because Ohio law does not require such an action.

 

— Green Party Observer     Montgomery County Recount

 

Morgan County(   )    **        No random sample

"The [Board of Elections] ladies said they chose the precinct to be hand counted, because it was just over three percent. It was not randomly chosen."

"When we asked to see the provisional and absentee ballots, we were told that there was no way to tell which they were, as they had been mixed back into their precincts. Once the envelopes and stubs were removed, there was no way to identify them any more."

"As for the provisional and absentee ballots that were not counted, (30 out of 160 total), they were not on hand. We were told they were locked away, and would be destroyed 60 days after the election."

"That said, it is my opinion... that the B of E in Morgan County ... did their best, as they understood the rules, to accomodate the recount process and the observation."

— Libertarian Observer    Morgan County Recount

(apparently there was no analysis of voter logs vs official votes or of undervotes and overvotes)

Morrow County (Punch Cards, Triad Tabulator  )     **   No random sample

"They picked Harmony Township to do the hand count on because it has 517 ballots cast. “

"At this point Observer S. spoke up and said that the precinct had to be selected at random. He explained that he felt that random was something like putting 36 precincts' numbers in a bowl and pulling one, if it was short, we would have to pull another one."

"This was not well taken. They said that they felt how they selected was random and that the board had the right to choose. They showed us in writing where they were getting this, and of course it is all in the interpretation of 'random.'"

"While this discussion was going on, one of the women left the room and came back and said that she had called the Secretary of State [Kenneth Blackwell] and that they were right."

"At this point I called you and discussed what was going on. Then I called the legal team and reported it."

"The hand count for Harmony Township continued."

  The first hand count did not match the machine count, so the ballots were further evaluated for overvotes and undervotes until a match was declared.   

"... All in all, this was very educational, and everyone seemed to want the same thing. I felt no ill intent at any time."

— Green Party Observer
Morrow County
Recount

(apparently no inspection of voter logs vs official votes, or of undervotes, overvotes, and provisionals was done)

Noble County  (  )   **     No random sample                 (small county-6,500 voters)

 

"the Board had selected the 27th Precinct for the recount. I objected on the basis that the board must randomly select the precinct(s) to be used for the recount ... I suggested a drawing by lot to determine the targeted precinct. Several board members and both employees argues that their method respected the rules ... We moved on."

The Director said that they had run a test deck manually, then a machine check on December 8, 2004

"The full run of the precincts proceeded on the machine. The recount was the same as the November 2nd count."

We then moved to the Poll Book/Reconciliation step and with the exception of one vote discrepancy in two precincts and a two vote discrepancy in one precinct, every precinct matched the November 2, 2004 count.

"We moved to the provisional ballots numbering 75, of which only 2 were not counted. We examined the provisional ballot envelopes and both were clearly received after the November 2, 2004 cut-off date."

— Green Party Observer                 Noble County Recount

 

Ottawa County  ( )   ?

Report by Green Party Observer

The Ottawa County recount started and ended on Monday, December 13th. They had two counting stations and we had two witnesses to observe the recount. There were also observers from the Democratic and Republican parties.

While I observed one of the Republican members of the board clearly advocate for one of his party's witnesses, who did not have the required documentation, there were no other incidents observed or reported.

I acted as the county Coordinator. The three percent analysis resulted in no discrepancies, so the rest of the county was completed by machine

(no indication as to whether sample was random; apparently no analysis of voter logs, provisionals, absentees, undervotes, overvotes)

Pauding County (  )      random sample, no discrepiencies

".. the three percent [of ballots chosen for the recount] was selected at random."

"[The witnesses] were ... allowed to make suggestions about which precincts to pick, and all the suggested names were put into a box and picked at random."

The three percent analysis resulted in no discrepancies, so the rest of the county was completed by machine.

The three witnesses here reported that they felt that the process was very professional in this county, and that the three percent was selected at random.

— Green Party Observer            Paulding County Recount

(no indication that voter logs, provisionals, absentees, overvotes, undervotes were checked)

 

Portage County   ( )

The 3% Hand count precincts, although not truly ascertained completely randomly, were chosen randomly of those that fit population numbers to get to the 3%, and came from random demographics representing both Dem wins and Rep wins.

We were given hours until completion to complete a full inspection of all rejected provisional and absentee ballots, allowing us to manually copy addresses of all those rejected in wrong precinct and for no signature. It was obvious by the notes on these ballots, that Portage county board employees and officials went out of their way to look up wrong addresses written on ballots by the voters, in both their system, and the Haines directory to find these folks, and count their votes. They sent cards to address listed on absentee ballots filed by voter incorrectly, and had those requests attached to each ballot.

The under votes we were asked to look for, were seen up close in the hand count as the voters choice NOT to vote for anyone for president. There were votes for local issues on these ballots, and the numbers were few and fell well within normal range of voting patterns.

At the end of the evening, the numbers on the tabulators screens, matched the hand count, which matched the final report sent to the tabulating computer. We took the numbers off the screens of the counting machines, as they ran each precinct in Portage County through. Precinct by precinct we compared those totals to election nights results in a long well witnessed process. There was one ballot that was soiled in this process, and they immediately re-made it to the best of their ability in plain site.

I have been trying to get access to the voter logs since Nov. 7th as a volunteer for Ohio Vigilance, but am confident after yesterday, talking to both Board of elections full and part time, bi-partisan staff, they just want to finish processing them to complete count, and do data entry into the voters files.  Although we have yet to count the signature books, we do have full access to them starting tomorrow. Portage County BOE Director Lois Enlow, will begin a rotating schedule for our volunteers counting the signature books starting tomorrow. I will add those numbers to my report as they come in for the entire county

"Portage County gets 2 thumbs up and an A+. Please tell everyone to maintain faith in the process... some Boards do have their act more than together!"

— Green Party Observer
Portage County
Recount

 

Putnam County   ( )       ***     County stated they did their own recount and there was no recount with observers present

"When I first called them, they had stated that their board would meet on December 15 to schedule their recount, but when contacted on December 15 to determine the results of that meeting, they stated that the recount had been conducted and completed that day."

— Green Party Observer
Putnam County
Recount

(there was no vote audit done for this county)

Sandusky County  (Optical Scan, ES&S  )        **       [some overvotes not counted(by machine) had clear voter intent.  Nader was not official candidate but received votes.    Some “overvotes” were not counted due to punch for Nader and another candidate-even though Nader was not supposed to be on the ballot.  There were 96 provisionals rejected.    some provisional ballots were not counted because of not having a signature on the outside envelope. Has there been a rule change on this?  There was also some talk of people being purged from the voting rolls if they hadn't voted recently so provisionals  rejected.. There were considerable overvotes. Probably some voters thought they had to vote separately for President and Vice-President, and were filling in the two ovals in the margin closest to their selections.  Some “overvotes” had vote for a candidate and candidate written in-clear intent so these were counted in the samples-sticker placed over write-in.   Interestingly, the two counting machines were ES&S, and even more interesting, there was a memo lying in plain sight on one of the tables with an ES&S letterhead. The memo gave a phone number to call if there were any problems on Novemeber 2, and requested that the election results be called in as soon as possible. Why would they need to know what the results were right away? The machines stored their information on a computer floppy disk drive built into the machine, but printed out the totals on an Okidata dot matrix printer on a table right by each machine. The hand counted precincts had the totals printed at the machine during the 3% test. During the full recount, the printers kept a running total of how many ballots they had counted and each time the counting was stopped when an uncountable ballot was encountered, but they did not print out any totals for the candidates as the recount progressed. I noted that to the person running the machine, and she said yeah, if it did that everyone would crowd around whenever it printed anything. Imagine that! So then I started to think, maybe there is a way this thing can be hacked after all. But I didn't find any proof of that.]

Details

Precinct summary sheets were given to assorted people in the room and 3 were told to pick a precinct. I chose Clyde — C, another chose Ballville Twp. A and another chose Rice Twp.   Not truly a random sample of all precincts.

The handcount and machine counts did not perfectly match but after examination of overvotes did match.    The rest of the precincts were then counted by machine.

I then proceeded to request the actual Polling Books and started checking those. Observer A. joined me shortly after.

Absentee List not counted and Provisional List not counted.

I was one of the witnesses of the recount in Sandusky County for the "GLib" party candidates on December 15. The Director of the Board of Elections, Barbara Tuckerman (D), told us at the outset that we were not to interfere with the count or stop it in any way. If we had questions that were not answered to our satisfaction we were to make note of them and relay them to our people later. They had prepared a test deck of ballots with a few each of minor party candidates, several each of the major candidates. What I didn't understand was the proportionally much larger stack of under- and over-voted ballots. When I asked why the stack was so large, it was explained to me but I still didn't get it, and I asked for clarification. I still didn't understand, but didn't want to look like an idiot or like I was trying to put a halt to things, so I made a note of it and hushed up. As near as I could tell, it looked like they were including one under-vote and one over-vote per precinct, or something like that. I didn't write down the number so I can't say exactly how many, but it was somewhere in the neighborhood of a hundred and quite disproportionate to number of legitimate test ballots in the deck. After hand counting these ballots and machine counting them, they reconciled the numbers and declared that the machines tested OK and we could proceed with the actual 3% machine vs. hand count test.

Earlier in the morning, I had observed that lists were being given to certain people to choose the precinct they wanted counted. This did not appear right to me, as it was my impression that the selection was to be by a scientifically valid random sampling plan. The scheme they were using certainly wasn't random. After talking this over with the other two GLib observers present, I recommended to them that we ask for a precinct as well. I believe it was Karen Wagner who made the selection, City of Clyde C. The other precincts that were chosen were Ballville TWP A CV, and Rice TWP. I paid close attention to Clyde C. The hand count matched the machine count after it was discovered that a ballot with a name written in on the write-in line was actually an under-vote by the machine count because the voter had neglected to fill in the bubble in the margin beside the name they had written. We were told we could go to lunch and the full machine recount would begin at one o'clock. People were visibly relieved that the machine count could go forward. I overheard some talk to the effect "we already agreed we weren't going to count all these by hand," which demonstrates overt collusion between Democrat and Republican factions.

We returned about 1:10 and found the machine count underway. I stationed myself at the same machine that had counted our precinct in the morning, and kept a close eye on what was going on. The machine would stop periodically with a message that a ballot had been found with no vote or too many votes on it, and these were set aside to be judged. One scenario was that there were some ballots where a candidate's bubble was filled in, as well as the one directly below it on the last line, that for a write-in candidate. The ruling was made that since the line for write-in candidate had been left blank, a white-out sticker would be placed over that bubble so the machine could count it as a vote for the candidate. Due to the rotation of names between precincts county-wide, ballots that were over-voted in this manner would go to various candidates, and if the voter filled in two or more bubbles higher up on the ballot, it was definitely an over-vote. Probably some voters thought they had to vote separately for President and Vice-President, and were filling in the two ovals in the margin closest to their selections. But none of the VP candidates had separate bubbles so there were some spoiled ballots on that score. If a voter had clearly made an effort to erase a bubble and filled in a different one but the machine counted it as an over-vote, it received a white-out sticker on the erased bubble so as to be properly machine counted. A couple of ballots that were machine read as under-votes had all the ovals on the ballot checked instead of filled in, so when this was shown to the judges the decision was made to fill in the bubble where the checkmark was. This all seemed quite fair to me and I was impressed at how well the Democrats and Republicans got along, joking about how much it hurt to repair those types of ballots for the other side, all in good fun. It was surprising to find ballots with most or all of the Presidential candidates' ovals filled in, or with none filled in at all. Puzzling, but relatively rare, ballots like these that registered no vote for President.

I asked for a copy of the official, certified results. I was told I would have to wait until they were done counting them. I said, no, I wanted the results that they had already sent in to the Secretary of State previously, those results had to be on record somewhere in the building. I was told OK, I could have a copy of that. I said, thank you. Can I have them now, please? I was told everyone was busy, I would have to wait until they were done counting. I said I really needed to have them before the recount was finished, that it was important to me that I have them now. I was polite but firm, and said there were plenty of people just standing around, and I am sure someone could make a copy of the results that had been sent to the SOS for me. So after about the fifth request it was delivered to me, and I thanked them graciously. The certified result for Clyde C was Bush 226. Kerry 212. The hand and machine recount was Bush 226, Kerry 214. I saw how this had happened by means of two of the over-voted ballots being corrected as previously mentioned, and I began to gain a little more confidence that perhaps the vote count was on the square in this county, but I couldn't swear to it.

I was asked why the recount was requested by the minor party candidates, and said I didn't know for sure. I said there has been a lot of reports of problems with the election, but I thought that it was in the larger counties and those with touch-screen voting. A man said something to the effect that any county that had those ought to smash them, they were ripe for fraud. I asked if I could quote him on that and he said yes, and I got his name. It was Tom Yonkers.

There are a couple of things about the spoiled ballots that bothers me. One is that in cases where the voter filled in a bubble, then crossed it out and filled in a different one the ballot was deemed an over-vote and no vote for President was recorded. I said it seems that the intent of the voter can be determined, and was told, no, you can't count those. I didn't argue the matter, but felt more strongly that this determination was improper when I later found out that the precincts were furnished with pencils that had no erasers. Also, it is my understanding that Nader was not allowed to be counted on the ballot, even though his name appears on it. In some cases Nader's bubble was filled in as well as that of another candidate. If a Nader vote did not count, shouldn't the voter's other choice (Kerry or whoever) be counted?

There was also some talk of people being purged from the voting rolls if they hadn't voted recently. If they showed up to vote and their name wasn't on the list but they insisted they had voted before and were registered, they were given a provisional ballot. If their name had been purged, however, I don't think their vote was counted. Is that right? Also, some provisional ballots were not counted because of not having a signature on the outside envelope. Was there a rule change on that? There were 713 provisional ballots issued, and 617 were ruled valid to count at some previous time. We were told the provisional ballots judged valid were included among those we were recounting.

Interestingly, the two counting machines were ES&S, and even more interesting, there was a memo lying in plain sight on one of the tables with an ES&S letterhead. The memo gave a phone number to call if there were any problems on Novemeber 2, and requested that the election results be called in as soon as possible. Why would they need to know what the results were right away? The machines stored their information on a computer floppy disk drive built into the machine, but printed out the totals on an Okidata dot matrix printer on a table right by each machine. The hand counted precincts had the totals printed at the machine during the 3% test. During the full recount, the printers kept a running total of how many ballots they had counted and each time the counting was stopped when an uncountable ballot was encountered, but they did not print out any totals for the candidates as the recount progressed. I noted that to the person running the machine, and she said yeah, if it did that everyone would crowd around whenever it printed anything. Imagine that! So then I started to think, maybe there is a way this thing can be hacked after all. But I didn't find any proof of that. The machines did not appear to have telephone lines hooked to them. When the recount was finished, each machine printed a total for the candidates that it had counted in that run, and I made notes of those results. The director took the floppies out of the machines and left the room, saying we should get the final results in about a half an hour. I saw the screen on the computer she was using a little later, it had a heading on it that said, "Unity Election Reporting Manager." I did a rough calculation, adding the numbers most recently printed together with the ones from the 3 precincts we had counted in the morning, and the numbers were relatively close. After everything was said and done, Bush picked up 26 votes, Kerry picked up 33. The election officials appeared to have done their jobs well, and the added votes were fairly allocated.

Results: Badnarik Bush Kerry Nader Peroutka Write-in over under totals

11/17     53 16,195 12,653      107   48 ? 225 139 29,451

12/15     53 16,221 12,686      107  48 31 173 136 29,455

The average voter turnout was 74.75%. Ballville F reported the highest, 87.58% turnout. It went Bush over Kerry 255-152. The lowest turnout was City of Fremont 4-E, 54.48%, which went Kerry over Bush 109-98. It appears that many of the precincts with above average turnout went Bush and those with below average turnout went Kerry, especially in Fremont. This could indicate tampering or suppression, or just a lower percentage of turnout by Kerry supporters. However, people most often turn out to voice their opinion to change things, not to make them stay the same. I do wonder what is going on.

For the most part, the election officials in Sandusky County were courteous and professional, and I thanked them for hard work on such a difficult day.

Summit County  ( )

Sample was preselected by Board with board IT staff present and 6 precincts chosen in a random manner.  

We were not able to closely observe the recount process, only from afar.   There has been a voter record lockdown.

A few of us are able to talk quietly near the front of the room and request signature poll books, precinct totals, and other reports that we want to inspect.  

"[The Summit County Deputy Director] was as antagonistic as he could be, as though his job was to make the day as difficult as possible; he asked us to stand away from the tables so we took a position at the end of the table. He drew a line on the floor with tape even though the volunteers were behind it the whole time. He even threatened to throw me out because he said I was too close..."

Another observer commented that one of the workers was shielding her count from the observers in what could be perceived as deliberate. "Team #7, Bernie, kept the ballots hidden by holding her contrast (yellow) sheet over them so I couldn't see the holes. She held the yellow sheet in her left hand and as she moved one ballot from in front of her with her right and into the line of sight, she covered it with the yellow sheet...."

"I was also threatened with removal from the premises if my lips moved, the deputy director was very determined that there was not to be a single whisper in the room from anyone..."

"When there was a discrepancy, it was nearly impossible to hear what the line supervisor determined the ballot should be counted as since voices were kept to a minimum by the employees and we were not allowed to lean in to hear what the discrepancy was. This happened a number of times throughout the recount process."

I requested the opportunity to review the rejected provisional and absentee ballots and was denied that request. I also requested that we view the signature books that corresponded to the ballots being hand counted, and again the request was denied. I did have a conversation later in the day with the director, Bryan Williams, and was told that the BOE was unable to accommodate our request today, and they would be happy to allow us to view these materials over the next few days.

The machine tally for the rest of the ballots in Summit County will take place on Thursday, December 16, beginning at 8:00 a.m. We are told that the process will take a few hours. 


We did receive the provisional ballot report; though I have no access to how to contact these folks. There is no problem receiving a voter database from the county, however, they withold phone numbers, so I will determine tomorrow how we want to proceed in contacting provisional rejected votes. Many of the provisional votes were accepted in Summit county. There are a few dozen that were tossed out for wrong precinct, or that the voter was not registered.

****************************************************************************************

Trumbull County               No report filed yet:

A careful review of the absentee vote in one Ohio county revealed that many more absentee votes were cast than there were absentee voters identified.      All absentee voters must be identified as such by name and residence in the

precinct poll books of the precinct in which they are registered. Over 100  precinct poll books in Trumbull County were checked for absentee voters and that   number of actual absentee voters was compared to the certified number of  absentee votes. There was an inflated difference in nearly every precinct of the   five communities examined. The five communities whose poll books were   carefully inspected for an absentee vote overcount are: Warren City (311), Howland  TownshipThe 106 precincts of these five Ohio communities, about 39% of all  precincts in Trumbull County, netted a total of 580 absentee votes for which there were no absentee voters identified in the poll books.      Assuming the same rate of inflated absentees statewide would produce a swing of 62,000 votes.

_http://www.scoop.co.nz/mason/stories/WO0412/S00154.htm_

http://blog.democrats.com/ohio-absentee

 

******************************************************

Tuscarawas County ( )    (no report completed)


"Two precincts were chosen to generate the necessary three percent." 

"Recount was suspended due to repeated discrepancy of 1 vote between hand count and machine... The tabulator continued to record one more vote for Kerry than the hand count."    With the Kerry candidate located in the 10 position, it would be possible that an overvote in the nine position would be determined to be an overvote during manual examination, but the tabulator would be programmed to recognize the 9 spot(Nader) to be vacant on the face card, and thus be recorded as a vote.

"The recount will be resumed Monday 12/20."    No further information on whether there was a recount of this county or if voter logs, provisionals, absentees,

undervotes, and overvotes were checked.

— Green Party Observer   Tuscarawas County Recount

 

Union County  (Punch Cards, Triad Tabulator )    **   no random sample

"[A Union County Board member] told us the day after the election that the hard drive on the machine, a Triad systems machine, failed and had to be replaced. They took the old hard drive with them when they replaced it but he sent a subpoena to them and got the old hard drive with a statement it had not been tampered with."  No indication of how the vote count was affected.

The director had picked precincts Marysville 11 and 13, and Allen 1.

A check of the voter logs for 2 of the precincts matched the vote totals except there were 7 Presidential undervotes in one and 2 in the other.

— Green Party Observer    Union County Recount

Van Wert County (Punch Cards, Triad Tabulator )       random sample , Triad serviced tabulators remotely before recount. 

Overall, the Board of Elections of Van Wert County was very cooperative."

The director explained that the precincts to be counted (Jackson and Van Wert 1D) were chosen at random at www.random.org"

"The hand recount of Jackson went smoothly and matched the total from election night. The Van Wert 1D count did not match the first time through. It was counted about three times and then matched.              

"When asked if Triad had serviced the machine, the deputy director and a board member stated that they had serviced the machine over the phone via modem on December 9th."

"... the director showed us the poll books for the two counted precincts ..."

"There were 295 total provisional ballots in the county, 23 rejected."    "The director provided us a list of the 23 provisional voters whose ballots were rejected and the reason why each was rejected. The list does not give an address or phone number for the voters."

"When asked if it was possible the precincts on the envelope were wrong, the director stated that the BoE had problems with that in the past and now instructed the poll workers to call the BoE if they ran out of provisional ballots and not to take from another precinct."

[Editor's note: This rule apparently was ignored in some counties, such as Montgomery.]

— Green Party Observer              Van Wert County Recount

 

Vinton County   ( Punch Cards, Triad Tabulator)   **  No random sample

"I arrived at the Vinton County BoE at approximately 11:30 am on Friday December 17. I had been told twice earlier on that week that everything would begin at 12 noon. However when I arrived, the Triad worker was already in the process of starting up the counting machine..."

"12:15 PM: ... I asked what precinct they would be using to calculate the three percent and how they arrived at that decision. I was told that the board had voted on it the days before and that it was picked because, at 277 ballots cast, it added up the closest to the three percent."

"I respectfully disagreed and requested that the three percent be chosen again [that is, randomly]. However, the motion was denied and the recount continued."

"The board counts at a comfortable speed and willingly holds up ballots so I can closely observe the punchcards"

"One overvote is challenged by Loretta, the Democratic party witness. The ballot has been punched once for Bush and once for Nader. As Nader is not on the Ohio ballot, the board thinks that the vote should count for Bush."

"Loretta points out that there weren't enough signs explaining to people that Nader was not on the ballot and someone may have voted for him anyway. The board votes on the matter and unanimously concludes that the vote be counted for Bush."

"The other ballot has been punched once for Bush and once in the wrong spot, electing no one. This is also placed in the votes for Bush pile."

"Just after 1:00 PM: The board selects the four precincts used to test the validity of the machine. This is done by writing the names of the precincts on paper and then randomly selecting from a cardboard box."

[Note: At 12;15 PM, the observer specifically asked for a random selection of votes to be recounted, and that request was denied. Later, at 1:00, precincts were randomly selected for a test of the machine — but not for the recount of the vote.]

At this point I am finally able to see the pollbooks. I randomly choose four precincts to examine. Crystal and Nell show them to me. I am able to see that all votes are accounted for and the signatures are clearly explained to me. On two occasions, the numbers on the inside of the book didn't match the number recorded. Nell and I were able to figure out what had happened. Everything was legit.

"I feel it important to recount that everyone present that day was a pleasure to work with. There were no problems at all and the people at the BoE were respectful of my needs."

— Green Party Observer  Vinton County Recount

Warren County   ( )    **     BOE stated that random sample was chosen without recount observer participation.  Audit was not allowed.

The sample precincts to be used for the recount were not chosen in the presence of the observers, but had been chosen prior to our arrival"

The ballots were in place on the counting tables when observer teams were allowed to enter the counting room (this seems to be in violation of ORC 3515.04 which states 'At the time and place fixed for making a recount, the board of elections, in the presence of all witnesses who may be in atendance, shall open the sealed containers containing the ballots to be recounted, and shall recount them.'

"... As stated in the original report, the observers from the Green and Democratic party questioned how those particular precincts were chosen 'at random' for the recount and one Board member said the numbers for those precincts were chosen 'out of a hat'. Subsequently, this was found out not to be how those precincts were chosen..."    "Why the board member stated that these precincts were 'selected from a hat' when knowing otherwise... does not make good sense to me"    The Board members did not tell the truth.

 "... Observers did not witness [the] test of the machine's accuracy."

"Contrary to what we expected (or wanted), observers did confirm via their inspection of ballots during the hand count that there were indeed ballots with NO choice for president."

Again, we are still awaiting a response from the BoE (who is waiting on a ruling from the Asst. Prosecutor) regarding our request for access to the poll books, rejected provisional and absentee ballots, and ballot accounting sheets. The review of these documents is necessary to make sure all the records are in order and legitimate votes haven't been rejected or that duplicate or prohibited votes have been counted. This is one area where we felt that the Republican Board members and the SoS representative, Christian Lobb, who had previously concurred with our request during a 'casual' conversation (until, it seems, the local Republican observers had some words with him), were particularly obstructionist, dismissive and condescending, as if what we were asking to do was both unnecessary and ridiculous, even though it is spelled out in the SoS Directive.

At least it seems to us to be clearly spelled out in the Directive. The SoS representative had told us that (between our first conversation with him and our 'official request') he had called the SoS's office for clarification and was told that the Directive was an 'outline', a 'guideline', and was 'optional'. It seemed to some of us (Greens and Dems) that the SoS rep. and Board (the strong personalities seemed to be the Republicans) inconsistently applied the SoS Directive, stating (or inferring) it was binding when it furthered their end and that it was 'optional' when it was something that we requested that they didn't want to do.

I heard the Green Party witness, Witness J., ask Susan Johnson about the provisional ballots. Susan indicated that there was no way to verify the rejected provisional ballots because there was no longer a name or address associated with each prov. ballot — they were simply in a stack.

"... we are still awaiting a response from the BoE (who is waiting on a ruling from the Asst. Prosecutor) regarding our request for access to the poll books, rejected provisional and absentee ballots, and ballot accounting sheets. The review of these documents is necessary to make sure all the records are in order and legitimate votes haven't been rejected or that duplicate or prohibited votes have been counted."

"To the best of our knowledge, (and that of the Asst. Director), the result of the recount was not certified."

— Green Party Observer   Warren County Recount

We specified that we wished to audit only a few precincts at random."

"Our request to perform these tasks was not accepted by the Board. Member Spaeth moved that our request should be reviewed by the County Prosecutor. His motion was seconded and passed unanimously by polling the members."

"Mr. Lobb said that the Secretary of State's 'Outline of Recount Procedures' was optional and that local Boards of Election could follow them or not."

— Green Party Observer B       Warren County Recount

 

Washington County (Optical Scan, )    **   No random sample

"The ballots were brought out from the ballot room and removed from a large wire cart to a table, spread out to keep precincts separate."

"Dorothy Kemp was explaining that they found 2 combinations of 2 precincts that would come closest to the 3%. We were unable to obtain any explanation as to how the decision would be random."

As we had been requested to view poll books first, Bill and myself asked about seeing them. We were told they were in the election office and we could adjourn there to view them if we wanted to. They cleared the room (I stayed till last to be sure) and locked the door and had a deputy sit outside the door.

The hand counts did not initially match the machine counts but overvotes and undervotes were marked up with a marker and stickers to “reflect the will of the voter” and then there was a match.  The Green observer disagreed that one overvote should have been counted for Bush since 2 candidates ovals were darkened.     A full machine recount was then carried out.  

After the counting was completed, they were told to save the results to disk. I asked for a printout of the totals from the counting machines. Mr. Walker asked me why I would want that when it would be the same as the printout from the BOE. I said that I wanted it anyway. At that point he noticed #6 ballots on top of machine #1 that had not been counted. He told me he couldn't give me a printout until he found out what to do with these 6 ballots. He said that he was going to ask Becky whether she wanted to add them in on her computer or have them do it at the counting station. (Good to know it can be done on her computer).

He returned, saying Becky told him to run them through the machines. He handed the worker at machine #2 a disk and told her to put that in and count the 6 ballots to add to it. I thought that it would just add 6 more and then save them to the disk so I was not concerned. They took the disk back and I again requested my printout. The poll worker hit "clear" on the machine and that was that. No data to print. I got a printout from machine #1 only. Then Mr. Walker came back from Becky Kirkbride's office with a disk marked "Machine #1 and #2". He said that when they entered the disk back in before it added all the votes from machine #2 again plus the #6. So they had to clear the machine (which had already been done by accident) and add the data (which at this point was the combination of both machines) then add the 6 ballots. I got a printout of this, but it is not raw data from the counting machines. (I feel like a failure on that one). The printout I received gives candidate totals but not totals of overvotes, undervotes or ovals marked for "write in."

— Green Party Observer   Washington County Recount

(no indication that voter logs, provisionals, absentees, overvotes, undervotes were checked)

Williams County           ( )    ?  

 "The three percent analysis resulted in no discrepancies"

"... the folks here did a very good job and were very pleasant"

— Green Party Observer     Williams County Recount

(no indication of whether sample was random;  no indication that voter logs, provisionals, absentees, overvotes, undervotes were checked; no details of recount given)

Wood County  ( )   ?

The Wood County recount started and ended on Monday, December 13th. They had two counting stations and we had two witnesses.

After the count I spoke by phone with both J and the other witness and both reported that everything went quickly and efficiently and that they did not complete any incident reports. The three percent analysis resulted in no discrepancies, so the rest of the county was completed by machine.

(no indication of whether sample was random, no indication that voter logs, provisionals, absentees, overvotes, undervotes were checked)

Wyandot County   ( )         random sample,  provisionals not recounted

The precinct was chosen randomly."

"Eleven absentee ballots were not counted because they arrived at the BOE after November 2, even though they were postmarked before election day..."

"Some points to ponder: In Wyandot County in 2003, the total number of precincts was reduced from 40 to 24, possibly causing some confusion about where to vote."     "Provisional ballots were not counted today but were on the table..."

"I think the BOE did an excellent job assisting us..."

— Green Party Observer   Wyandot County Recount

(no indication that voter logs, provisionals, absentees, overvotes, undervotes were checked)

Ohio Counties

Reports pending

·                          Adams  Ashtabula  Athens  Belmont  Carroll  Clark  Crawford  Darke    Erie    Franklin    Geauga    Harrison  Highland  Jackson  Lawrence

·                          Logan  Madison  Muskingum  Perry   Pickaway  Pike  Preble  Richland  Ross  Scioto  Seneca  Shelby  Stark  Trumbull




In Perry County, we inspected the voter logbooks, which showed 100 people voting in one precinct without any signatures, leaving no way to verify who actually voted. A lawsuit asking the courts to overturn the results of the Perry County auditor's race alleges that the number of votes exceeded the number of people who signed the voting books.

Under Ohio law, each county must randomly choose a precinct to recount by hand and by machine. If the two counts do not match, officials must conduct a countywide recount by hand. Most county Boards of Elections, however, chose to preselect the sample precinct, a violation of the law. Some counties refused to proceed with a full hand recount when the hand and machine tallies failed to match.

In two of the three counties we observed, technicians from Diebold and Triad, manufacturers of voting machines and vote-counting software, were present during the entire recount. The Diebold technician in Hardin County was actively involved in giving instructions to the observers. Further, he arrived the day before the recount to prepare the machines and data disks that contain the election results.

At an Ohio hearing convened by Rep. John Conyers, D-Mich., an affidavit was filed on behalf of Sherole Eaton, an election worker in Hocking County, describing how an employee of Triad may have tampered with the vote tabulator when he dismantled it three days before the Dec. 13 recount. In her affidavit, she states the technician told her "how to post a 'cheat sheet' on the wall so the ... count would come out perfect and we wouldn't have to do a full hand recount of the county."

Conyers has asked the FBI to investigate. Attorneys for John Kerry filed two motions on Monday to preserve the evidence in this case and to take the deposition of the Triad technician.

********************************************

On Dec. 16 we attended a Franklin County Board of Elections public hearing in Columbus. Citizens expressed their anger and outrage at having to wait up to seven hours to vote. One woman referred to the long lines as a "new poll tax." The Election Protection Coalition reported that of the 464 complaints about long lines in Ohio, 400 came from Columbus and Cleveland, where a large proportion of the state's Democratic voters live.

Ohio Secretary of State Kenneth Blackwell, who co-chaired the Bush-Cheney campaign in Ohio, has come under heavy criticism for his handling of the election and the recount. He refused to respond to questions by several members of Congress. Conyers charged in a subsequent letter that Blackwell's refusal to answer questions is "part of a pattern of decisions that have worked to obstruct and stonewall a search for the truth about voting irregularities." Blackwell is also seeking a protective order to keep him from being interviewed as part of a court challenge to the Ohio election. We will never have a clear picture of Ohio's election results because of the lack of a statewide manual recount, lack of a voter-verified paper trail for many of the state's voters who used electronic voting machines, questions of possible machine tampering, and untold numbers of discouraged voters deterred by long lines. We call on Sen. Mark Dayton to join Rep. Maxine Waters and other members of Congress to stop the approval of the Electoral College votes on Jan. 6 until there is a full investigation into what really happened in Ohio.

Mark Halvorson is a co-founder of Citizens Alliance for Secure Elections - Minnesota, a grass-roots group that advocates for election integrity. Kirk Lund is an attorney.

***************************************************************************

In my opinion, there should have been fair state-wide rules about how certain types of "spoiled" ballots would be counted in the recount.

The county I helped with had optical scan ballots. In cases where it was obvious that a voter bubbled in a choice by mistake, then made an attempt to erase it and filled in another choice, the ballot was counted as legitimate. However, when instead of making an attempt to erase, the filled in bubble was X'd out and another filled in, these ballots were rejected, going into the "overvote" pile. I objected to that practice since it appeared the voter's intent was clear, but I was overruled. (I later found out the voters were supplied with pencils without erasers attached)

In cases where a voter voted for both Nader and another candidate, these also went into the spoiled, overvote stack. But since Nader was not supposed to be on the ballot at all, these ballots should have counted for whoever the voter picked besides Nader. If these two types of ballots were consistently counted statewide, it might have made a difference in the outcome, IMO.

Yes, I don't think the people who paid for the recount got their money's worth. The recount, like the election, was fixed, rigged, hacked and stolen six ways from Sunday.

"Bruce" <hallbr01@holmes.ipfw.edu>

**** · KNOX COUNTY -- (1) Mostly white precincts received more voting machines than precincts with a majority of African-American voters; and (2) certain precincts lacked enough voting machines.

· HAMILTON COUNTY -- (1) Mostly white precincts received more voting machines than predominantly African-American precincts; (2) there were some voting machine errors keeping voters from inserting ballots ""all the way into certain machines……""; (3) candidates Kerry-Edwards names were omitted from some absentee ballots; and (4) a Republican precinct judge was asking every voter for their address and ""being a jerk about it.""

· JEFFERSON COUNTY -- ""some challenged voters were not notified that their registration was challenged and their right to vote was in question. Their names were merely published in a nearly unreadable list in the local newspaper.""

· LAKE COUNTY -- ""some voters received a memo on bogus Board of Elections letterhead informing voters who registered through Democratic and NAACP drives that they could not vote.""

·

· MAHONING COUNTY -- (1) voting machine errors included one precinct ""recorded a negative 25 million votes""; (2) twenty to thirty ""ES&S iVotronic machines needed to be recalibrated during the voting process because some votes for a candidate were being counted for that candidate''s opponent""; (3) about a dozen of these machines had to be reset because they ""essentially froze."" And, (4) ""there were numerous reported instances of vote hopping (in which a voter selecting Kerry for President saw the choice displayed on the machine ''hop'' to Bush for President).""

· MERCER COUNTY(Pa) -- Voting machine errors showed ""289 people cast (punch card) ballots, but only 51 votes were recorded for president. The county''s Web site appeared to show a similar conflict…… It would appear that about 4,000 votes (nearly 7%) were not counted for a candidate.""

· MIAMI COUNTY - (1) In Concord Southwest precinct ""voter turnout was a highly suspect and improbable 98.55%. In Concord South precinct, there was a highly improbable 94.27% voter turnout."" (2) After all precincts had reported, ""18,615 votes came in"" and are ""statistically suspicious..""

· MONTGOMERY COUNTY - Voting machine errors showed two precincts with ""25% presidential undervotes. This means no presidential vote was recorded on ? of the ballots. The overall undervote rate for the county was 2%. The undercount amounted to 2.8 percent of the ballots in the 231 precincts that supported candidate Kerry, but only 1.6 percent of those cast in the 354 precincts that supported candidate Bush.""

· SANDUSKY COUNTY - (1) There was an overcount ""when a computer disk containing votes was accidentally inserted into the vote tabulating machines twice by an election worker."" (2) Also, it was ""discovered some ballots in nine precincts were counted twice.""

· STARK COUNTY - ""The Election Board rejected provisional ballots cast at the wrong precinct in the right polling place."" This had been allowed in other elections.

· TRUMBULL COUNTY - a voter forged a registered voter''s name and gave a different address. When the registered voter arrived, she was allowed to vote.

· WARREN COUNTY - ""there were irregularities in the county on election night when officials locked down the county administration building and blocked anyone from observing the vote count.""

· IN SEVERAL COUNTIES, besides Franklin and Mahoning, ""there were numerous reported instances of vote hopping.."" Kerry voters saw their votes go to Bush.

The attorneys are basing their arguments on Ohio''s Revised Code 3515.16, entitled ""Testimony in Supreme Court."" All testimony will be given in the form of depositions.

The contestors may have 20 days to take and file testimony and the contestees 20 more days, but ""the court may render such judgments and make such orders as the law and facts warrant.""

Michael I. Shamos from the Institute for Software Research International, at Carnegie Mellon University, believes ""it is possible to overturn the Elector''s vote to match the will of the people if the irregularities are enough to overcome the margin of victory.""

**************************************************